TABLE 2.

Summary table of included studies: supplementation¹

Study	Age: duration	Iron type	Intervention, n	Control, n	Total, n	Other study arms, n	Effect size	P value	Subgroups of note	Detrimental effect of intervention
Abdelrazik et al. (38); 2007; India	6 mo: 12 mo	43 mg ferrous gluconate	198	50	348		Proportion: intervention 75.8%, placebo 50%	P = 0.03	Of the group that received iron, those with normal ferritin at baseline had higher rates of diarrhea; P = 0.04	Yes
Baqui et al. (36); 2003; Bangladesh	6 mo: 6 mo	20 mg ferrous sulfate	165	157	476	154 MMN	Adjusted odds: iron 1.01(0.91–1.13), MMN 1.15 (1.02–1.29), control 1.0	P < 0.05	A zinc and zinc + iron arm was excluded from extraction; MMN was not well tolerated with a 41% drop out rate; iron alone had little effect on diarrhea incidence, but MMN containing iron had a significant impact	Yes
Chang et al. (34); 2010; Bangladesh	6–18 mo: 6 mo	6.25 mg “Iron”	201	201	799	199 iron and zinc, 198 zinc only	Incidence rate: iron 2.7, placebo 2.3, combined iron and zinc 2.0, iron and zinc alternative days 2.1, zinc alone 2.3.	P < 0.05	Iron alone significantly increases risk of diarrhea in children; this effect is mitigated by the addition of zinc; giving iron to underweight children had less detrimental effects on diarrhea than those who were of normal weight.	Yes
Chen et al. (71); 2013; China	3–6 y: 6 mo	1–2 mg/kg ferrous sulfate	98	104	292	90 vitamin A + iron	Incidence rate: iron 0.4, placebo 0.43, vitamin A + iron 0.28	P > 0.05	Significant decrease when iron is combined with vitamin A P < 0.05	No
Dewey et al. (28); 2002; Honduras Cohort	4 mo: 3 or 5 mo	1 mg/kg ferrous sulfate	36/40 (3/5 mo)	42	118		Proportion (over whole study duration): intervention 64%/58% placebo 50%		Iron supplementation reduced the risk of diarrhea among infants with Hb < 110 g/L at 4 mo, but led to an increase in diarrhea among infants with Hb > 110 g/L at 4 mo P = 0.03; NB: combined cohorts used: Sweden/Honduras; morbidity data specific to 4–6 mo and 6–9 mo reported, but overall morbidity incidence extracted only	Yes
Dewey et al. (28); 2002; Swedish Cohort	4 mo: 3 or 5 mo	1 mg/kg ferrous sulfate	30/30 (3/5 mo)	36	96		Proportion (over whole study duration): intervention 27%/30% placebo 14%		Combined data, as above	Yes
Luabeya et al. (32); 2007; South Africa	6 mo: 18 mo	10 mg ferrous fumarate	109	113	335	113 zinc + vitamin A	Number of cases: intervention 89 placebo 98 vitamin A + zinc 92	P = 0.484		No
Mitra et al. (35); 1997; Bangladesh	29 mo: 15 mo	125 mg ferrous gluconate	118	131	249		Diarrhea episodes per child per year: intervention 2.8 (1.6–4.8), control 2.5 (1.6–5.0) dysentery episodes per child per year: intervention 2.5 (0.9–4.8), control 2.5 (0.9–4.8)	Diarrhea P = 0.32, dysentery P = 0.84	Dysentery episodes per child per year, under 12 months of age: iron 5.2 (2.4–7.8), control 3.5 (2.1–4.8); P = 0.03	Yes
Richard et al. (72); 2006; Peru	0–4 y (multiple strata): 7 mo	15 mg ferrous sulfate	60	61	187	66 zinc	Risk ratio: intervention 0.97 (0.78–1.21), iron + zinc 0.89 (0.70–1.12), control 1.0.	P = 0.32		No
Rosado and Allen (73); 1997: Mexico	1.5–3 y: 12 mo	20 mg ferrous sulfate	54	56	165	55 iron + zinc	Episodes per year: intervention 76, iron + zinc 46, placebo 62	P > 0.05 intervention P < 0.05 iron + zinc		No

¹Hb, hemoglobin; MMN, multiple micronutrient.